How Predictive Models Are Rewriting the Congenital Syphilis Story – Healthcare Blog

Author: Keira Kelly

Each semester I have the privilege of mentoring nursing students in their maternal and infant clinical work. As the semester begins, their enthusiasm is contagious. They shared stories of witnessing first births, helping new mothers breastfeed, and performing developmental assessments on pediatric patients. As the semester progressed, I saw their demeanor change. “You're right, we took care of it. other Today is a baby with congenital syphilis. ” Their reflections on the clinical day were a mixture of emotions: frustration, anger and sadness, as they watched their fragile babies battle an infection that no child should have to endure.

When I first tell my nursing students that they may be caring for babies with syphilis during their clinical rotations, they look at me with wide eyes in disbelief. “Didn’t we cure syphilis in the 1950s?” someone asked. Some of my students usually remember hearing about the Tuskegee Study, but most don't know that we are still fighting (and failing) against congenital syphilis in the United States today.

Congenital syphilis occurs when a mother passes the infection to her baby during pregnancy or delivery. This condition is almost entirely preventable with prompt screening and treatment, but the number of cases continues to increase at an alarming rate. Congenital syphilis cases in the United States increased by 183% between 2018 and 2022, from 1,328 to 3,769 cases. This national trend is also reflected at the state level, with Texas reporting 179 cases in 2017 and 922 in 2022. Over these five years, the rate of babies born with congenital syphilis in Texas rose from 46.9 to 236.6 per 100,000 live births, a dramatic increase that requires action.

Despite having one of the most comprehensive prenatal screening laws, Texas currently has one of the highest rates of congenital syphilis in the nation. According to the Texas Department of State Health Services, policy requires syphilis screening at three points during pregnancy:

(1) First prenatal check-up

(2) Late pregnancy (but no earlier than 28 weeks)

(3) Upon delivery

But here's the question: What happens if women never get prenatal care? How do we reach people who have never set foot in an OB/GYN office during pregnancy? Screening laws protect only those who have access to care. In 2022, more than 1/3 of Texas mothers whose babies were diagnosed with congenital syphilis did not receive any prenatal care. Each of these cases represents a failure of our current healthcare system, which is supposed to protect the most vulnerable, but still fails to reach those most in need.

Socioeconomic and systemic barriers often limit access to health care for vulnerable groups and communities. Congenital syphilis disproportionately affects infants born to mothers with limited access to health care, unstable housing, poverty, maternal drug use, and inadequate prenatal care. Many women also avoid or delay prenatal care because of stigma, fear of judgment from health care providers, or fear of getting tested for substance abuse.

Imagine if we no longer relied solely on women attending prenatal appointments for screening, but could identify who was most likely to have a baby with congenital syphilis whenever they interacted with any part of the health care system. By leveraging existing electronic health record (EHR) data and artificial intelligence (AI), we can build predictive models that predict maternal and infant health outcomes.

These models can include things like prenatal care utilization, zip codes, and other clinical data. Patients flagged as high risk in the EHR can automatically trigger nurse navigation referrals for further evaluation and care coordination. Rather than limiting syphilis screening to obstetrical visits, this approach could identify high-risk patients at any point of contact: emergency department, primary care, behavioral health, substance use treatment, or community outreach clinics.

Predictive models have proven successful in improving other clinical outcomes such as sepsis, diabetes, and even preterm birth. We already have the EHR system and the required data. We just need to develop and apply the model. These success stories demonstrate that, through data analytics and artificial intelligence, improving outcomes for congenital syphilis is not only possible, but achievable.

Currently, both U.S. and Texas policies focus on syphilis screening requirements during prenatal care. But what about women who have never participated in traditional prenatal care? How do we protect babies from congenital syphilis? We must critically evaluate our approaches and develop policies that are commensurate with the realities of today's health care system.

Many pregnant women seek care at emergency rooms or urgent care clinics for unrelated problems such as urinary tract infections, fever, or coughs. Each encounter provides an opportunity for health care providers to intervene and prevent congenital syphilis transmission. Policies should be updated to require screening of pregnant women who do not meet existing screening guidelines at every health care visit and ensure that pregnant women identified as high risk are followed within 48 hours.

Once high-risk patients are identified through predictive models, geographic mapping can help public health professionals effectively target outreach efforts. The tool creates visual maps that reveal infection clusters and highlight hot spots where testing, education and community resources should be focused. Health departments often use this approach to allocate resources to where they are needed most.

Funding to build predictive models and integrate them into EHR systems may come from state and public health grants. Once developed, the ongoing costs of maintaining this model will be minimal compared with the rising costs of congenital syphilis. The average hospital cost for an infant born with congenital syphilis is approximately $56,802, nearly four times higher than for an infant without congenital syphilis. Preventing even a small number of cases could quickly offset the investment costs required to develop and implement the model.

The dramatic increase in congenital syphilis cases represents a failure of our health system, a failure defined by missed opportunities for prevention. While AI will never replace the human element of compassionate care, it can give us the data we need to make a lasting impact on vulnerable populations and improve maternal and child health outcomes.

Remaining stagnant under our current ineffective policies borders on negligence. Having technology available and not using it is a failure of rescue in many ways. But a combination of technology and compassion can change the outcome of this story. I thought of the students' faces, the frustration and doubt in their eyes. I wish I could tell them that this will be the last time they see a baby with congenital syphilis, but unless things change, this is just the beginning.

Kayla Kelly, MSN, RN, CPN is a nursing instructor and doctoral candidate at the University of Texas at Tyler